With the development of information office, the information needs to be acquired or extracted from printed applications and saved in the background system to be processed further when the printed applications are received or involved in data collection processes in the enterprises. The APPLICATION represents a form of interaction of a user with an entity in which the entity requests data in a structured format and the user provides said requested data in said structured format. An application may contain one or many forms, as well as other documents/information. In addition, the printed application refers to a printed matter using paper as the carrier of the information in the application. The current techniques used to capture the data from the printed applications go from 2D barcodes to OCR (Optical Character Recognition), and also manually keying the data as a last resource. These three methods above depend highly on the print quality of the paper document received. In addition, their cost is high, labor intensive, and the two latter methods (OCR and Manual Type In) are prone to getting errors easily.
Therefore, how to improve the quality of the data acquisition from printed applications becomes an important problem that needs to be solved.